Intercom’s MCP server lets an agent read your support history — conversations, contacts, companies and Help Center articles — and write in two careful ways: articles and internal notes. It cannot send a reply a customer would see.
https://mcp.intercom.com/mcp
Claude Sonnet 4.5
Chat with 60+ AI models on the same workflow — switch to a different model mid-conversation and re-run the same prompt, or use Compare mode to put several side-by-side and balance quality vs. cost.
OAuth (recommended) or an Intercom access token as a bearer token. US and EU workspaces only; Australian workspaces are not yet supported.
How models use it and what it is built for.
The server exposes 14 tools. Most are read-only: search and fetch for a general lookup, then search_conversations, get_conversation, search_contacts, get_contact, list_companies, get_company, list_articles, search_articles and get_article. Three write: create_article and update_article for the Help Center, and add_internal_note, which leaves a note only teammates can see. That boundary is the point. An agent can research a ticket, find similar past conversations, check the customer’s company and draft an answer as an internal note, but a person still sends the reply. It is also a quick way to find the questions your Help Center does not answer yet.
Typical tools an AI model can call. Exact names vary by version.
Taken from the official Intercom documentation — see Intercom MCP documentation for the full reference.
Examples
US
https://mcp.intercom.com/mcpEU
https://mcp.eu.intercom.com/mcpFor workspaces hosted in the EU.
Legacy SSE (US)
https://mcp.intercom.com/sseDeprecated. Use only for clients without Streamable HTTP.
Copy any of these into MCP Agent Studio after connecting.
Find open conversations this week that mention refunds, and summarise the common causes.
Pull up everything we know about the customer in conversation 1234, including their company.
Which questions came up more than five times this month with no Help Center article covering them?
Draft a reply to conversation 1234 as an internal note, citing the relevant article.
This is not a single-model product: you get the same MCP connection with 60+ models (Claude, GPT, Gemini, DeepSeek, open-weight, and more), you can switch mid-conversation, and you can open Compare mode to run the same prompt against multiple models at once. The card above is a suggested starting point for this server — not the only choice.
Default pick for Intercom
Claude Sonnet 4.5
Support threads are long and full of back-and-forth. Sonnet 4.5 summarises them faithfully and keeps the customer’s actual problem separate from the noise.
Listing tools proves the server is reachable, not that a model can work with it. Evals go further: they read every tool on the server, write a test suite from its real schemas, and run it — code decides pass/fail on the responses (schema conformance, error codes, pagination, result caps) while a scoring model grades plain-English tasks driven through the tools.
Get a pass/fail report per tool with the evidence behind each verdict — and replay the same suite after every schema change. Destructive tools are excluded from the run.
Run evalsOpen MCP Agent Studio with the connection pre-filled. Add your token, pick any of 60+ models, and start chatting — no install required.
Open Agent StudioCommon questions about connecting, scoping and using it safely.
No. add_internal_note leaves a note only teammates can see. The server cannot send customer-visible replies, so a person always sends the answer.
Help Center articles, with create_article and update_article, and internal notes on conversations. Everything else is read-only.
US and EU workspaces are supported, at mcp.intercom.com and mcp.eu.intercom.com. Australian workspaces are not yet supported.
Read and list users and companies, read conversations, write conversations (for internal notes only), and read and write articles.
Use /mcp, which is Streamable HTTP. The /sse endpoint is deprecated and kept for older clients.
HubSpot
Query CRM, deals and tickets — and draft follow-ups with full record context.
Salesforce
Query CRM records, update pipeline, and draft outreach in natural language.
Slack
Search workspace, read threads, post updates and work with Slack from chat.
Notion
Search, summarise and edit your Notion workspace with AI.